NUMERICAL ANALYSIS OF THE BED MORPHOLOGY IN THE REACH BETWEEN CABRUTA AND CAICARA IN ORINOCO RIVER.

Transcription

1 NUMERICAL ANALYSIS OF THE BED MORPHOLOGY IN THE REACH BETWEEN CABRUTA AND CAICARA IN ORINOCO RIVER. Raul A CABRITA F MEE13634 Supervisor: Shinji EGASHIRA ABSTRACT The present study aims to evaluate numerically the river bed scouring resulting from sand bars formation and migration, and channel plan forms such as a change of flow width, curved flows, in addition to local scouring due to construction of bridge pier. To achieve the objective of this study, the numerical model International River Interface Cooperative (iric) developed by Hiroshi Takebayashi et. al (2009) is employed to simulate and predict the morphological changes in river bed, and empirical formulas are applied to estimate the local scour depth at bridge piers. (1) The river bed variation is predicted well in terms of numerical model proposed by Takebayashi etal. The bed erosion along the outer bank increases with time in initial stage, and then oscillates temporally. The maximum depth increases with flow discharge. (2) The local scour depth resulting from construction of bridge pier is estimated in terms of several empirical methods. The results suggest that Andru s method is suitable for study area. (3) Combining the results of bed deformation and local scour due to bridge piers can be useful for the estimation of the depth of bridge footing. Keyword: Orinoco River, Bed morphology, Bridge, Local scours, Piers. 1. INTRODUCTION The National Railway System of Venezuela (NRSV) is a main concern for the nation and is currently under construction. The Instituto de Ferrocarril del Estado (IFE) is the authority on all issues and has planned the completion of the NRSV for 20 years. One of the line of the NRSV is Chaguaramas - Cabruta Caicara with a total length of 201 km that would facilitate communications between the central and southern of Venezuela. The connection between Chaguaramas and Caicara depends on the construction of the third bridge over the Orinoco River. The global statistics of bridges over waterways show that most bridges fail due to hydraulic reasons such as erosion around piers or abutments. So, it is important to evaluate erosion depth resulting from morphological changes and local scour at bridge piers. This study aims to evaluate the river bed scouring resulting from sand bars formation and migration, and channel plan forms such as a change of flow width, curved flows, in addition to local scouring due to construction of bridge pier. To achieve the main objective, the numerical model proposed by Hiroshi Takebayashi et. al (2009) is employed to simulate and predict the morphological changes in river bed, and empirical formulas are applied to estimate the local scour depth at bridge piers. 2. REVIEW OF LOCAL SCOUR DUE TO BRIDGE PIERS The local scour at bridge piers is a subject that has concerned the attention of many researchers for many years. This is due to erosion resulting from the interaction between local flow and sediment transportation. Table 2.1 shows three methodologies which are composed of parameters associated with Universidad Central de Venezuela, Venezuela. ICHARM The International Centre for Water Hazard, Japan. 1

2 flow discharge, sediment transportation and geometry of the bridge piers to estimate the maximum possible local scour depth at piers. Researchers Table 2.1 Methodologies for estimation maximum possible local scour depth at piers Formulas Mazza Alvarez and Echavarria Alfaro Andru Nakagawa & Suzuki d s = 2.4K s K θ b Where, d s is the maximum possible depth scours at piers [m], K s and K θ are the shape and alignment factor, respectively, and b is the Piers width [m] d s y = 0.80 Where, d s is the maximum possible depth scours at piers [m] and y is the water depth [m] d s b = log 10 ( b ) d 50 Where, d s is the maximum possible depth scours at piers [m], b is the pier diameter [m], and d 50 is the size of the particle sediment. 3. CHARACTERISTIC OF CHANNEL MORPHOLOGY (2.1) (2.2) (2.3) 3.1. Description of the area of study Location of the bridge Figure 3.1 shows the study area in which the third bridge will be constructed over Orinoco River. The reach between Cabruta and Caircara in Orinoco River are specified as follows: bed elevation ranges between m and the m; an average discharge is 32,970 m 3 /s and velocity is 1.28 m/s; median grain size diameters are 0.26 mm to 0.54 mm; a mean concentration of suspended sediments is 38mg/l. In addition, the ratio of fall velocity of the sediment and the shear stress velocity (w 0/u *) is For the analysis of the channel morphology of the reach between Cabruta and Caircara del Orinoco in Orinoco River, we need river cross section data, discharge data, water level data, grain size distribution data, channel roughness, etc. For this study, the basic data was obtained from different sources and organization Figure 3.1 Location of the study area. (Source: Google Eath) 3.2. Data analysis Figure 3.2 shows the cross section of interest at which the 3rd bridge will be constructed, and Figure 3.3 shows the monthly flow discharge at Caicara gauge station of Orinoco River, respectively. The width of this section is about 2,600 m and the elevations ranges from 8.70 m to m. The maximum discharge is 77,900 m 3 /s and the minimum is 2,135 m 3 /s in this reach. Usually, the maximum level takes place from around the end of July to the beginning of August. 2

3 Elevation [m] Discharge [m 3 /s] Percent finer than indicated size [%] Figure 3.4 shows the particle size distribution curves in the area of Cabruta. The median diameters for these samples were 0.26 mm at station 3, 0.50 mm at station 5, and 0.54 mm at station 9. The sample at station 3 consisted mostly of fine to medium sand; 100 percent of the material is finer than 1.00 mm and about 3 percent of the material is finer than sand (<0.062 mm) Diameter [mm] Station 1 Station 3 Station 5 Station 9 Station 13 Station 17 Station 21 Station 25 Figure 3.4 Particle size distribution of bed material Orinoco River Distance [m] Figure 3.2 cross section of interest at which the 3rd bridge will be constructed Time [month] Figure 3.3 Monthly discharge of Orinoco River at Caicara Gauge Station Jan to Dec NUMERICAL METHOD TO PREDICT MORPHOLOGICAL CHANGES 4.1. Governing equations for the numerical model The governing equations, which are depth integrated two dimensional forms are employed in the numerical model. These are composed of mass conservation equation for water flow, suspended sediment of flow body and bed sediment as well as of momentum conservation equation for water flow. Mass conservation equation for flow body (Continuity equation) h t + x (uh) + y (vh) = 0 (4.1) Momentum conservation equation for flow body uh t + x (uuh) + y (uvh) = gh x (h + z b) τ x ρ + x (hσ xx) + y (hτ (4.2) yx) vh t + x (uvh) + y (vvh) = gh x (h + z b) τ y ρ + y (hσ yy) + x (hτ yx) (4.3) u τ x = τ b (4.4) u 2 + v 2 3

4 v τ y = τ b (4.5) u 2 + v 2 Mass conservation equation of suspended sediment. ch t + ruch x + rvch y = x (hε c x x ) + y (hε c y y ) + E D (4.6) Mass conservation of bed sediment z b t λ ( q bx x + q by y + E D) = 0 (4.7) Where, h is the flow depth. u, are depth- averaged flow velocity along the longitudinal and transversal direction, respectively, t is the time, g is the acceleration due to gravity, z b, and ρ are the bed elevation and mass density of water, respectively. σ xx, σ yy, τ yx, and τ xy are the Reynolds stresses, τ b is bed shear stress, and D are erosion and deposition rate of suspended sediment, respectively. q b, c, r, and λ are bed load rate, concentration of sediment, factor correction, and porosity of bed sediment, respectively. ε x, and ε y are the x and y components of dispersion coefficient 4.2. Grid generation for numerical computations and boundary conditions. Figure 4.1 shows the study area in terms of grids generated in general coordinate system. A uniform artificial channel, which is 10 km long and constant bed slope (I= 0.04 ), is attach to both side of the study reach in order to conduct a numerical computation stable. Numerical computations are conducted with the following boundary conditions. At upstream boundary condition is considered a uniform steady flow, the equilibrium sediment transport rate for both bed load Q b and suspended load Q s. At downstream of the study area there is no information about water surface elevation. So, for downstream boundary condition is considered the normal depth. Upstream Study area Downstrea m Figure 4.1 Physical plane of study area in terms of the generated grids 5. MORPHOLOGICAL CHANGES IN STUDY REACH In this chapter shows the results and discussion of the case where is considered a constant discharge of 32,920 m 3 /s that is corresponding to the average discharge of all the historical data since Figure 5.1 and Figure 5.2 shows the bed shear stress distribution and the bed elevation change of the river after 20 days of computation, respectively. In this figures suggest as the bed shear stress increase, the bed erosion happens, on the contrary, when the shear stress decrease the sediment deposition take place. In this channel the maximum bed shear stress occurs: reach between the sections 29 and 26 near to the right bank, expansion (between sections 20 and 18) and contraction in the channel (sections 12 to 10). In those areas also take place the maximum erosion. On the other hand, the maximum deposition take place in areas as: left bank between sections 29 and 26, right bank between sections 25 and 23; also, in this areas occurs the minimum bed shear stress in the channel. Moreover, the value obtained for the cumulative transported suspended sediment is 3.96 x 10 7 ton, value that is bigger than the value obtained in previous study. The reason of this differences are the condition established in the numerical model as: bed material type (uniform) and flow discharge (steady flow). 4

5 Figure 5.1 Bed shear stress of the Study Area for 20days computation Figure 5.2 Elevation change of the Study Area for 20days computation Figure 5.3 shows the results of the maximum possible local scour at bridge pier in meters for seven different methodolgies. It can be appreciate that Andru s method provided the maximum scour depth at pier and only depend the water depth. The minimum scour depth is obtained with Yaroslavtziev s method. The values obtained using the methodologies of FHWA and Breuser still are low in one case and in the other only depend of the geometry of the pier. Laursen and Toch, Mazza, and Nakagawa and Susuki present similar values Figure 5.3 Maximum possible local scour depth at piers in meters For each case that was considered in this research is important to understand the variation in time of the local scour depth at bridge piers. Figure 5.4 shows the variation in time obtained after applied the equation (2.2) corresponding to Andru s method. When is considered the flow discharge of 8,400 m 3 /s for 20 days of computation, the maximum local scour (6.89 m) occurs after five days and then start to decrease to a depth of 4.89 m. For a flow discharge of 32,920 m 3 /s the local scour is around 13 m and do not present great variation along the time. In the case of 77,909 m 3 /s only was obtained computation for 3 days and the behavior is similar as the two other cases. According to the previous explanation the main reason of the variation of this local scour in each case is due to the change in the bed elevation of the river. In the case with a flow discharge of 8,400 m 3 /s is where presented more variation in time. 5

6 Scour depth (ds) [m] Time [day] ds [m] (Q=8,400m3/s) ds [m] (Q=77,909 m3/s) ds [m] (Q=32,920 m3/s) Figure 5.4 Variation in time of the local scour depth at bridge pier 6. CONCLUSIONS The depth of bridge pier footing should be determined taking the local scouring and river bed evolution into consideration. The results obtained from( present study are summarized as follow: (1) The river bed variation is predicted well in terms of numerical model proposed by Takebayashi etal. The bed erosion along the outer bank increases with time in initial stage, and then oscillates temporally. The maximum depth increases with flow discharge. (2) The local scour depth resulting from construction of bridge pier is estimated in terms of several empirical methods. The results suggest that Andru s method is suitable for study area. (3) Combining the results of bed deformation and local scour due to bridge piers can be useful for the estimation of the depth of bridge footing.. ACKNOWLEDGEMENTS I would like to express my appreciation and gratitude to my Sensei Shinji EGASHIRA for his valuable advice to complete this study. Also, I wish to express my special gratitude to Ms. Yuri MEDINA and Mr. Jacinto ARTIGAS, for their support. REFERENCES (1) Egashira, Shinji (2009). Mechanics of sediment transportation and river changes, Lectures Note (2) Zanabria, Luis (2011). Hidraulica fluvial, Lectures Notes. (3) Nordin, C. and Perez, D. Sand Waves, Bars, and Wind Blown Sand of the Orinoco River, Venezuela and Colombia.United States Geological Survey Water Supply with collaboration of Ministerio del Ambiente y Recursos Renovables of Venezuela (4) Ven Te Chow (1994). Open Channel Hydraulics, Mc Graw Hill. (5) Takebayashi, H. (2011), Morpho 2D. Solver Manual. (6) Instituto de Ferrocarriles del Estado. Sistema Nacional de Ferrocarril de Venezuela. Dic 2013 (7) Ashida, K. and Michiue, M., 1972, Study on hydraulic resistance and bed-load transport rate in alluvial streams, Proc. of JSCE, No. 206, pp (8) Ashida, K. and Michiue, M., 1972, Study on hydraulic resistance and bed-load transport rate in alluvial streams, Proc. of JSCE, No. 206, pp (9) B. Y. Liu, 1991, Study on Sediment Transport and Bed Evolution in Compound Channels. Thesis presented to Kyoto University. 6